Original Article
Primary gastrointestinal non Hodgkin’s lymphoma chemotherapy alone an effective treatment modality: Experience from a single centre in India Raina Vinod, Sharma Atul, Vora Amish, Shukla NK*, Deo SVS*, Dawar R** Departments of Medical Oncology, *Surgical Oncology and **Pathology, Institute Rotary Cancer Hospital, All India Institute of Medical Sciences, New Delhi, India Correspondence to: Atul Sharma, E-mail:
[email protected]
Abstract BACKGROUND: Gastrointestinal tract (GI) is the most frequently involved extra nodal site in non-Hodgkin’s lymphoma (NHL). Surgery, radiotherapy and chemotherapy (CT) have been used mostly in various combinations, but lately chemotherapy alone has emerged as an effective option. The purpose of this study is to evaluate efficacy of CT alone in treatment of primary GI-NHL and to compare the results with combined CT + surgery. SETTING AND DESIGN: Retrospective analysis of case records of GI NHL patients. MATERIALS AND METHODS: Over a 15-year period (1986-2000), 77 new cases of primary GI-NHL were registered at our center. GI-NHL was defined according to standard criteria. All patients received chemotherapy. RESULTS: The median age was 32 years (Range 9-80). Endoscopy / CT guided biopsies were performed in 42% (32) of patients for the purpose of diagnosis. Laparotomy was done in 58% (45) of patients to establish a diagnosis or as primary or debulking treatment. Stomach and intestines were involved in 47% (36) and 53% (41) patients respectively. Early stage disease was present in 37% (29). Seventy eight percent of tumors were intermediate to high grade, 43% (33) received only CT while 57% (44) received CT + surgery. Five years EFS and OS were: 72% and 65% for all patients; 72% and 67% for CT only group; 60% and 64% for CT + surgery group (P=0.05). Four patients died of neutropenic infection. CONCLUSION: Organ preservation strategy using chemotherapy alone (CT) can be successfully employed in a significant number of patients with primary GI-NHL. Key Words: Primary non-Hodgkin’s lymphoma of gastrointestinal tract, chemotherapy.
Introduction The incidence of non-Hodgkin’s lymphoma (NHL) has been increasing over the last three decades. During the same period an increase in incidence of extra nodal NHL has also been noted. [1] Gastrointestinal tract represents the most frequent extra nodal site and it accounts for 4% of all gastrointestinal malignancies. Although many large series of primary GI-NHL describing patterns of presentation and outcome have been published the consensus over the ideal treatment 30 30 CMYK
for GI-NHL still remains the subject of debate. [2-10] Interpretation of the outcome data of these studies is hampered by differences in case selection, staging system, pathological classification and therapy. Surgery, radiotherapy (RT) and chemotherapy (CT) have all been used either alone or in combinations. Superiority of either single or combined modality over each others is still not proven. Surgery has traditionally remained the treatment of choice and chemotherapy is often used after surgery. Adjuvant radiotherapy is also practiced in some cases. Nevertheless surgery and radiotherapy are not without significant morbidity. NHL being a highly Indian Journal of Cancer | January–March 2006 | Volume 43 | Issue 1
Raina et al: Gastrointestinal non Hodgkin’s lymphoma chemotherapy
chemo sensitive disease it is now highly questionable whether radical or mutilating surgery is still necessary.
administered.[17] All patients who were operated outside were also given chemotherapy as above.
Advances in endoscopic techniques for obtaining tissue diagnosis, refinements in radiological methods such as CT scans and guided biopsies and efficacy of chemotherapy have created a new option in the treatment of GI-NHL. Chemotherapy has the advantage of organ preservation; in addition it is effective for micro-metastases and hence takes care of systemic disease. In the older studies of high grade GI-NHL, chemotherapy was employed either as an adjunct to surgery or in combination with radiotherapy. However more recent studies suggests that chemotherapy alone may be as effective particularly in primary gastric lymphomas.[11-13]
Response criteria and end points were reported according to published guidelines.[18] Re-evaluation after completion of treatment included endoscopy, CT-scans and complete blood count and biochemistry in addition to clinical examination. Patients were examined every three months for one year and six monthly thereafter.
Materials and Methods In this study we have analyzed 77 patients of primary GI-NHL treated at our center over the last 15 years (1986-2000). This review includes clinical features, histopathological classification, site of involvement, treatment outcome and prognostic factors. Some of our patients were operated in other hospitals before being referred to us. This gave us the opportunity to compare surgery + CT vs. CT alone group. Primary GI NHL was defined according to Lewin et al i.e. patients had to present with GI symptoms or have predominant lesions in the GI tract.[14] The initial evaluation of all patients included a complete history and physical examination, complete blood count, liver and renal function tests, chest X-ray and computed tomographic scans. All patients where primary surgery was not performed underwent upper gastrointestinal endoscopic or colonoscopic studies or guided biopsies for obtaining tissue diagnosis. Bone marrow aspiration and biopsy was done in all patients. Patients were staged according to the Ann-Arbor classification as modified by Musshoff.[15] The histopathology specimens of all patients were reviewed and classified according to the International Working Formulation, which was the classification followed in our institute during study period.[16] All patients, irrespective of stage were administered chemotherapy. For patients with diffuse large B cell lymphoma and lymphoma of indeterminate histology, six cycles of CHOP (Cyclophosphamide 750 mg/m2, doxorubicin 50 mg/m2 , Vincristine 1.4 mg/m2 with maximum of 2 mg, prednisolone 100mg for 5 days every 21 days) were given. For Burkitt’s lymphoma, eight cycles of dose intensive MCP 842 was Indian Journal of Cancer | January–March 2006 | Volume 43 | Issue 1
Statistical analysis
The date of analysis was July 1, 2002. Event free survival (EFS) was determined from the time patients entered into complete or partial remission until recurrence or death from any cause, overall survival (OS) as time from diagnosis until death from any cause or till the last follow up. Patients in remission or alive were censored at the last known date of follow up evaluation. OS and EFS were calculated using Kaplan Meier method. For univariate analysis, log-rank test was performed using SPSS 10.0 version. Results The clinical features and patient characteristics are given in [Table 1]. Pain was the commonest symptom (81% 64 patients); fever and weight loss were the other frequent symptoms, either of them being present in 54.5% (43) of patients. Forty eight percent of patients presented with nausea, vomiting, constipation associated with abdominal pain. Only 3% of patients presented with either hemetemesis or malena. Seventy eight percent (60) of patients had either intermediate or high grade NHL, none of our patients had low grade NHL. Four patients had Burkitt’s lymphoma. Seventeen patients (22%) could not be classified further into intermediate or high grade, as slides available were not of good quality. Stage wise, 16.8% (13) and 20.7% (16) of patients had early disease (i.e. stage I and stage Table 1: Clinical features and patient characteristics Median age (range in years)
32 (9-80)
Male to female ratio
2.2:1
Clinical features �
B- Symptoms present
43 (54.5%)
�
Median duration of symptoms
3 months
�
Pain
64 (81%)
�
Hemetemesis
2 (2.5%)
�
Malena
3 (3%)
�
Subacute intestinal obstruction
38 (48.1%) 31 CMYK31
Raina et al: Gastrointestinal non Hodgkin’s lymphoma chemotherapy
II 1 respectively), while 37.6% and 15.5% had disseminated disease i.e. stage II2 and stage IV respectively. Seven (9%) patients could not be staged in the absence of the complete information.
Table 3: Shows clinical features, histology type of lesion and stage wise distribution in GI lymphoma according to the site involved Features
Stomach
SI
L. Intestine
The diagnosis of lymphoma was established by endoscopy in 25(30.4%) patients, by USG/ CT guided biopsy techniques in 7 (8.9%) patients. Diagnostic laparotomy was performed in 45 (58.8%) patients. As shown in [Table 2], ulceroproliferative type of growth was seen in (45.4%) patients. Nodular, infiltrative and polypoidal pattern was seen in 14.2% and 9% of patients respectively and mixed pattern was noted in 16.8% of patients. Stomach was the most common organ involved (46.7%-36) as shown in [Figure 1]. Median age for small intestine NHL was much lower (30 years) compared to stomach and large intestinal NHL. B symptoms and abdominal pain was present in the majority of patients irrespective of site of involvement. There was equal distribution of patients as far as histology type of lesion and stage were concerned. Clinical features, histology, stage and site distribution has been shown in [Table 3].
· Number
36 (46.7)
28 (36.3)
13 (17)
· B-symptoms
22 (61)
16 (57.1)
5 (38.5)
Intermediate grade 14 (38.8)
12 (42.8)
5 (38.5)
High grade
10 (27.8)
10 (35.1)
2 (15.4)
Mixed
—
1 (3.6)
2 (15.4)
Burkitt’s
2 (5.6)
2 (7.1)
—
Unclassified
10 (27.8)
3 (10.7)
4 (30.8)
Stage I
7 (19.4)
5 (17.9)
1 (7.7)
Stage II1
9 (25)
5 (17.9)
2 (15.4)
Stage II2
11 (30.6)
11 (39.3)
7 (53.8)
Stage IV
6 (16.7)
5 (17.9)
1 (7.7)
The median follow up of the cohort was 72 months (range 0 to 112 months). Main side effects of chemotherapy were myelosuppression, vomiting and diarrhoea. No patient had perforation or hemorrhage following chemotherapy. Four patients died of neutropenic fever. The OS and EFS of the patients with primary GI-NHL were 65% and 72% respectively [Figures 2 and 3]. We had three groups of patients depending on the treatment they received; 1. Surgery (complete resection) + chemotherapy, 2. Surgery (partial resection) + chemotherapy, 3. Only chemotherapy. Survival of different treatment has been
Resection Complete
6 (16.7)
9 (32.1)
6 (46.2)
Incomplete
9 (25)
11 (39.3)
3 (23.1)
No surgery
21 (58.3)
8 (28)
4 (30.8)
Table 2: Type of lesion Type of lesion
N = 77 (%)
Ulceroproliferative
35 (45.4)
Nodular
11 (14.2)
Infiltrative
11 (14.2)
Polypoidal
07 (9)
Mixed
13 (16.8)
Histology
Stage wise*
Figurss in the parenthesis are in percentage
shown in Table 4. A total of 23 deaths were recorded during the study period. Eight patients died of disease progression after relapse, four of neutropenic infection, nine patients did not achieve CR and died, two died in CR due to causes unrelated to the disease. Patients who did not achieve CR were treated with salvage protocols like IMVP-16 (Ifosfamide, Methotrexate and VP-16) or MINE (Mitoxanterone, Ifosfamide and VP -16). Prognostic variables
We analyzed prognostic factors using univariate and multivariate analysis. We did not find any correlation between B symptoms, age, sex, stage, histopathological grade, site of involvement and survival. Patient who received chemotherapy alone had 5 year survival of 67% compared to 60-64% in group who had surgery and chemotherapy (P=0.05). Discussion
Figure 1: Shows distribution of GI-lymphoma according to site involved
32 32 CMYK
Primary GI-NHL represents a heterogeneous disease with regard to various characteristics like stage, site of involvement, histological subtypes and treatment offered. The commonest presenting symptom in GI-NHL is Indian Journal of Cancer | January–March 2006 | Volume 43 | Issue 1
Raina et al: Gastrointestinal non Hodgkin’s lymphoma chemotherapy Survival Function
Table 5: Sites of involvement in primary GI-NHL: different studies First author Raina et al (Present study)
No. of subjects
Distribution
77
36 stomach 28 small intestine 13 large intestine
Peter Koch[3]
371
277 stomach 32 small intestinal 26 theocecal region
no at risk 76 27 17
24 multiple sites 15
4
2
d’ Amore
1
[2]
306
175 stomach 09 intestine
Figure 2: Overall survival in months
22 both Survival Function
Radaskiewicz[9]
307
244 stomach
Gurney[22]
883
463 stomach
63 intestine 419 intestine Liang
[23]
433
238 stomach 184 intestine
Morton[8]
175
78 stomach 95 intestine
no at risk 57 25 19
15
3
2
MALT type.[3] In our series, there was no patient with MALT lymphoma.
1
Figure 3: Evemt free survival in months
Fifty two percent of our patients had disseminated disease at the time of diagnosis (stage II 2E + stage IVE) in contrast to the Western reports of 10% 31%.[7,3]
Table 4: EFS and OS of various treatment groups Type of treatment
N
EFS OS (60 months) (60 months)
Complete resection + CT
21
60%
64%
Partial resection + CT
23
62%
60%
Only chemotherapy
33
68%
67%
All group
77
72%
65%
abdominal pain.[7] In our series, 81% of patients had abdominal pain as presenting feature. More than half (54.5%) of our patients had B symptoms, which is higher than Western data.[3,19-21] We have noted equal distribution of gastric and intestinal lymphomas, Gastric: 46.7%, Intestinal 53.3%. [Table 5] summarizes anatomic location of the initial disease in the GI tract reported in recent studies.[2,3,8,9,22,23] In the majority of the studies, stomach is the commonest site of involvement. With the introduction of the entity MALT lymphoma by Issacson, more and more gastric MALT lymphomas are diagnosed. [24] As reported in recent series by Peter et al, 40% of stomach NHL were of Indian Journal of Cancer | January–March 2006 | Volume 43 | Issue 1
Surgery with or without chemoradiotherapy has been the mainstay in the treatment of GI-NHL but this may be questioned. In a report by Brigitte et al 90% of cases underwent surgery followed by chemotherapy with 61% OS at 2 years. [6] Two important large prospective studies have been reported recently. In German trial, of 371 patients with primary GI-NHL 44% had intermediate to high-grade lymphoma; received surgery + chemotherapy and had 5 year overall survival of 64%. [3] In a study on gastric lymphoma from Mexico 589 patients were randomized to either surgery (S) with or without radiotherapy (SRT) and chemotherapy (SCT) and chemotherapy (CT) alone, actuarial curves at 10 years showed that overall survivals (OS) were: S: 54%; SRT: 53%; SCT: 91%; CT: 96% (P
